Advertisement

Single-channel characterization of the chitooligosaccharide transporter chitoporin (SmChiP) from the opportunistic pathogen Serratia marcescens

Open AccessPublished:September 13, 2022DOI:https://doi.org/10.1016/j.jbc.2022.102487
      Serratia marcescens is an opportunistic pathogen that can utilize chitin as a carbon source, through its ability to produce chitin-degrading enzymes to digest chitin and membrane transporters to transport the degradation products (chitooligosaccharides) into the cells. Further characterization of these proteins is important to understand details of chitin metabolism. Here, we investigate the properties and function of the S. marcescens chitoporin, namely SmChiP, a chitooligosaccharide transporter. We show that SmChiP is a monomeric porin that forms a stable channel in artificial phospholipid membranes, with an average single-channel conductance of 0.5 ± 0.02 nS in 1 M KCl electrolyte. Additionally, we demonstrated that SmChiP allowed the passage of small molecules with a size exclusion limit of <300 Da and exhibited substrate specificity toward chitooligosaccharides, both in membrane and detergent-solubilized forms. We found that SmChiP interacted strongly with chitopentaose (Kd = 23 ± 2.0 μM) and chitohexaose (Kd = 17 ± 0.6 μM) but did not recognize nonchitose oligosaccharides (maltohexaose and cellohexaose). Given that S. marcescens can use chitin as a primary energy source, SmChiP may serve as a target for further development of nutrient-based antimicrobial therapies directed against multidrug antibiotic-resistant S. marcescens infections.

      Keywords

      Abbreviations:

      ChiPs (chitoporins), GlcNAc (N-acetylglucosamine), (GlcNAc)2 (chitobiose), (GlcNAc)n (n >2, chitooligosaccharides), IM (inner membrane), ITC (isothermal microcalorimetric), Kd (equilibrium dissociation constant (μM)), MM (M9 minimal medium), OM (outer membrane), SmChiP (Serratia marcescens chitoporin)
      Serratia marcescens is a facultative Gram-negative, soil-borne bacterium (
      • Mahlen S.D.
      Serratia infections: from military experiments to current practice.
      ) that frequently causes outbreaks of community- or hospital-acquired infections in adults and children (
      • Yoon H.J.
      • Choi J.Y.
      • Park Y.S.
      • Kim C.O.
      • Kim J.M.
      • Yong D.E.
      • et al.
      Outbreaks of Serratia marcescens bacteriuria in a neurosurgical intensive care unit of a tertiary care teaching hospital: a clinical, epidemiologic, and laboratory perspective.
      ,
      • Voelz A.
      • Müller A.
      • Gillen J.
      • Le C.
      • Dresbach T.
      • Engelhart S.
      • et al.
      Outbreaks of Serratia marcescens in neonatal and pediatric intensive care units: clinical aspects, risk factors and management.
      ). S. marcescens may cause respiratory infections (
      • Jones R.N.
      Microbial etiologies of hospital-acquired bacterial pneumonia and ventilator-associated bacterial pneumonia.
      ,
      • van der Vorm E.R.
      • Woldring-Zwaan C.
      Source, carriers, and management of a Serratia marcescens outbreak on a pulmonary unit.
      ), meningitis (
      • Aronson L.C.
      • Alderman S.I.
      The occurrence and bacteriological characteristics of S. marcescens from a case of meningitis.
      ,
      • Rabinowitz K.
      • Schiffrin R.
      A ward contamination by Serratia marcescens.
      ), and urinary tract infections (
      • Kawecki D.
      • Kwiatkowski A.
      • Sawicka-Grzelak A.
      • Durlik M.
      • Paczek L.
      • Chmura A.
      • et al.
      Urinary tract infections in the early posttransplant period after kidney transplantation: etiologic agents and their susceptibility.
      ). The microorganism is highly resistant to most antimicrobial agents (
      • Nikaido H.
      Outer membrane barrier as a mechanism of antimicrobial resistance.
      ,
      • Stock I.
      • Grueger T.
      • Wiedemann B.
      Natural antibiotic susceptibility of strains of Serratia marcescens and the S. liquefaciens complex: S. liquefaciens sensu stricto, S. proteamaculans and S. grimesii.
      ,
      • Arakawa Y.
      Systematic research to overcome newly emerged multidrug-resistant bacteria.
      ). Numerous strains of S. marcescens possess quorum-sensing regulated virulence R factors, which are endogenous plasmids that carry antibiotic-resistance genes (
      • Cooksey R.C.
      • Thorne G.M.
      • Farrar Jr., W.E.
      R factor-mediated antibiotic resistance in Serratia marcescens.
      ). Most strains also carry highly effective ABC-type efflux pumps (
      • Shirshikova T.V.
      • Sierra-Bakhshi C.G.
      • Kamaletdinova L.K.
      • Matrosova L.E.
      • Khabipova N.N.
      • Evtugyn V.G.
      • et al.
      The ABC-type efflux pump MacAB is involved in protection of Serratia marcescens against aminoglycoside antibiotics, polymyxins, and oxidative stress.
      ). In addition, the cell envelope of S. marcescens has a complex lipopolysaccharide layer, which increases the bacterial membrane barrier. S. marcescens also forms biofilms (
      • Hall-Stoodley L.
      • Costerton J.W.
      • Stoodley P.
      Bacterial biofilms: from the natural environment to infectious diseases.
      ,
      • Satpathy S.
      • Sen S.K.
      • Pattanaik S.
      • Raut S.
      Review on bacterial biofilm: an universal cause of contamination.
      ,
      • Srinivasan R.
      • Devi K.R.
      • Kannappan A.
      • Pandian S.K.
      • Ravi A.V.
      Piper betle and its bioactive metabolite phytol mitigates quorum sensing mediated virulence factors and biofilm of nosocomial pathogen Serratia marcescens in vitro.
      ,
      • Ray C.
      • Shenoy A.T.
      • Orihuela C.J.
      • González-Juarbe N.
      Killing of Serratia marcescens biofilms with chloramphenicol.
      ). Treatment of S. marcescens infections can be difficult, owing to its intrinsic resistance to most antibiotics, including amoxicillin, ampicillin, the first generation of cephalosporins and carbapenems, and some fluoroquinolones (
      • Stock I.
      • Grueger T.
      • Wiedemann B.
      Natural antibiotic susceptibility of strains of Serratia marcescens and the S. liquefaciens complex: S. liquefaciens sensu stricto, S. proteamaculans and S. grimesii.
      ,
      • Jacoby G.A.
      AmpC beta-lactamases.
      ,
      • Bes T.
      • Nagano D.
      • Martins R.
      • Marchi A.P.
      • Perdigão-Neto L.
      • Higashino H.
      • et al.
      Bloodstream infections caused by Klebsiella pneumoniae and Serratia marcescens isolates co-harboring NDM-1 and KPC-2.
      ). In severe cases, patients require intensive medical treatments using fourth-generation cephalosporins or piperacillin/tazoabactam (
      • Harris P.N.
      • Ferguson J.K.
      Antibiotic therapy for inducible AmpC beta-lactamase-producing Gram-negative bacilli: what are the alternatives to carbapenems, quinolones and aminoglycosides?.
      ).
      S. marcescens can utilize various types of organic material as its carbon source and can produce chitin-active enzymes to hydrolyze chitinous materials and use them as an energy source (
      • Vaaje-Kolstad G.
      • Horn S.J.
      • Sørlie M.
      • Eijsink V.G.H.
      The chitinolytic machinery of Serratia marcescens--a model system for enzymatic degradation of recalcitrant polysaccharides.
      ). Chitin breakdown is initiated by a chitin-active lytic polysaccharide monooxygenase that breaks the chitin polysaccharide chain into chitin fragments by oxidative cleavage (
      • Uchiyama T.
      • Kaneko R.
      • Yamaguchi J.
      • Inoue A.
      • Yanagida T.
      • Nikaidou N.
      • et al.
      Uptake of N,N′-diacetylchitobiose [(GlcNAc)2] via the phosphotransferase system is essential for chitinase production by Serratia marcescens 2170.
      ,
      • Suzuki K.
      • Shimizu M.
      • Sasaki N.
      • Ogawa C.
      • Minami H.
      • Sugimoto H.
      • et al.
      Regulation of the chitin degradation and utilization system by the ChiX small RNA in Serratia marcescens 2170.
      ). The oxidized chitin fragments are hydrolyzed further by chitinases (
      • Vaaje-Kolstad G.
      • Horn S.J.
      • Sørlie M.
      • Eijsink V.G.H.
      The chitinolytic machinery of Serratia marcescens--a model system for enzymatic degradation of recalcitrant polysaccharides.
      ,
      • Uchiyama T.
      • Kaneko R.
      • Yamaguchi J.
      • Inoue A.
      • Yanagida T.
      • Nikaidou N.
      • et al.
      Uptake of N,N′-diacetylchitobiose [(GlcNAc)2] via the phosphotransferase system is essential for chitinase production by Serratia marcescens 2170.
      ,
      • Monreal J.
      • Reese E.T.
      The chitinase of Serratia marcescens.
      ,
      • Watanabe T.
      • Kimura K.
      • Sumiya T.
      • Nikaidou N.
      • Suzuki K.
      • Suzuki M.
      • et al.
      Genetic analysis of the chitinase system of Serratia marcescens 2170.
      ,
      • Suzuki K.
      • Sugawara N.
      • Suzuki M.
      • Uchiyama T.
      • Katouno F.
      • Nikaidou N.
      • et al.
      Chitinases A, B, and C1 of Serratia marcescens 2170 produced by recombinant Escherichia coli: enzymatic properties and synergism on chitin degradation.
      ,
      • Takanao S.
      • Honma S.
      • Miura T.
      • Ogawa C.
      • Sugimoto H.
      • Suzuki K.
      • et al.
      Construction and basic characterization of deletion mutants of the genes involved in chitin utilization by Serratia marcescens 2170.
      ,
      • Synstad B.
      • Vaaje-Kolstad G.
      • Cederkvist F.H.
      • Saua S.F.
      • Horn S.J.
      • Eijsink V.G.H.
      • et al.
      Expression and characterization of endochitinase C from Serratia marcescens BJL200 and its purification by a one-step general chitinase purification method.
      ) to short-chain chitooligosaccharides, which are transported into the periplasm of the bacterial cell through outer membrane (OM) porins. Previous studies reported the existence of several porins in the OM of S. marcescens, including Omp1, Omp2, Omp3, OmpF, and OmpC. These general diffusion pores take up small, hydrophilic molecules by passive diffusion (
      • Sánchez L.
      • Ruiz N.
      • Leranoz S.
      • Viñas M.
      • Puig M.
      The role of outer membrane in Serratia marcescens intrinsic resistance to antibiotics.
      ,
      • Begic S.
      • Worobec E.A.
      Regulation of Serratia marcescens ompF and ompC porin genes in response to osmotic stress, salicylate, temperature and pH.
      ). On the other hand, a disaccharide (chitobiose) and higher molecular weight chitooligosaccharides, which cannot pass through the general diffusion pore, require a chitooligosaccharide-specific porin (ChiP) for their entry into the periplasm (
      • Monreal J.
      • Reese E.T.
      The chitinase of Serratia marcescens.
      ). Watanabe et al. (
      • Watanabe T.
      • Kimura K.
      • Sumiya T.
      • Nikaidou N.
      • Suzuki K.
      • Suzuki M.
      • et al.
      Genetic analysis of the chitinase system of Serratia marcescens 2170.
      ) previously identified the chiP gene encoding ChiP (later named SmChiP) as part of the chiPQ-ctb gene cluster in the genome of S. marcescens 2170. The expression of chiP mRNA is controlled by the nontranslated chiX small RNA that binds to the gene’s 5′-untranslated region, which contains a 17-nucleotide Shine-Dalgarno sequence preceding the chiPQ-ctb gene cluster. The same study also demonstrated that the ΔChiP mutant had a drastically reduced growth comparing to the growth of the wildtype, confirming that ChiP played an important role in cell survival. We previously identified and characterized a chitooligosaccharide-specific porin, namely SmChiP, from S. marcescens (
      • Amornloetwattana R.
      • Robinson R.C.
      • Soysa H.S.M.
      • van den Berg B.
      • Suginta W.
      Chitoporin from Serratia marcescens: recombinant expression, purification and crystallization.
      ). In this report, we employed time-resolved single channel electrophysiology to examine the ion-conducting properties and substrate specificity of SmChiP. We also carried out in vivo cell studies to demonstrate that S. marcescens can employ chitooligosaccharides as its primary carbon source.

      Results

      Sequence analysis and AlphaFold2 structural prediction

      We previously reported that some characterized chitoporins (ChiPs) exist as trimers and others as monomers, depending on the organism. For example, VfChiP identified from Vibrio furnisii (
      • Keyhani N.O.
      • Li X.B.
      • Roseman S.
      Chitin catabolism in the marine bacterium Vibrio furnissii. Identification and molecular cloning of a chitoporin.
      ), VhChiP from Vibrio campbellii (formerly classified as Vibrio harveyi) type strain ATCC BAA 1116 (
      • Suginta W.
      • Chumjan W.
      • Mahendran K.R.
      • Janning P.
      • Schulte A.
      • Winterhalter M.
      Molecular uptake of chitooligosaccharides through chitoporin from the marine bacterium Vibrio harveyi.
      ), and VcChiP from Vibrio cholera type strain O1 (
      • Soysa H.S.M.
      • Aunkham A.
      • Schulte A.
      • Suginta W.
      Single-channel properties, sugar specificity, and role of chitoporin in adaptive survival of Vibrio cholerae type strain O1.
      ) are each composed of three identical subunits, while EcChiP from Escherichia coli is a monomeric porin (
      • Soysa H.S.M.
      • Schulte A.
      • Suginta W.
      Functional analysis of an unusual porin-like channel that imports chitin for alternative carbon metabolism in Escherichia coli.
      ,
      • Soysa H.S.M.
      • Suginta W.
      • Moonsap W.
      • Smith M.F.
      Chitosugar translocation by an unexpressed monomeric protein channel.
      ).
      Figure 1 shows structure-based sequence alignment of monomeric ChiPs, including SmChiP and EcChiP, in comparison with trimeric ChiPs, including VhChiP and VcChiP. From the sequence alignment, SmChiP has high (77%) sequence identity with EcChiP, but low (17%) sequence identity with the trimeric ChiPs. The secondary structure elements of SmChiP are similar to those of EcChiP but different from those of the trimeric ChiPs. Overall, monomeric ChiPs have a single-barrel structure composed of 18 β-strands connected by eight extracellular loops, while VhChiP and other trimeric ChiPs contain 16 β-strands connected by eight extracellular loops. Loop L3 of trimeric ChiPs, which is typically identified as the pore-confining loop, is exceptionally long when compared with that of the monomeric ChiPs, while loops L2 and L6 of monomeric ChiPs are longer than those in the trimers. The N-terminal segment that is characteristic of trimeric ChiPs is longer than that of the monomeric ChiPs. Notably, the N-terminal segment of VhChiP contains a short helix that serves as the plug that regulates the open and closed states of the VhChiP pores (
      • Aunkham A.
      • Zahn M.
      • Kesireddy A.
      • Pothula K.R.
      • Schulte A.
      • Baslé A.
      • et al.
      Structural basis for chitin acquisition by marine Vibrio species.
      ). This N-plug is unique to VhChiP and is not present in other ChiPs (Fig. 1).
      Figure thumbnail gr1
      Figure 1Structure-based sequence alignment of four characterized chitoporins: SmChiP (protein id: A0A0PBS3), EcChiP (protein id: P75733), VcChiP (protein id: Q9KTD0), and VhChiP (protein id: L0RYU0). The amino acid sequences of these proteins were aligned by MAFFT alignment and displayed in Jalview v. 2.11.2.1. The 2D structural elements of monomeric SmChiP, displayed at the top, were constructed from AlphaFold2 (https://alphafold.ebi.ac.uk/), while the 2D structural elements of trimeric VhChiP, displayed at the bottom, were constructed from the crystal structure of the native form (PDB id: 5MDQ). Helices are represented by cylinders, β-strands by thick arrows, and loops by lines. SmChiP, Serratia marcescens chitoporin.
      Figure 2 shows the AlphaFold2-predicted structure of SmChiP, compared with the crystal structure of the native VhChiP (PDB id: 5MDQ) (
      • Aunkham A.
      • Suginta W.
      Probing the physiological roles of the extracellular loops of chitoporin from Vibrio campbellii.
      ). Figure 2A shows the side view (left panel) and top view (right panel) of VhChiP, consisting of three identical trimers, each of which contains the characteristic N-terminal plug, 19 amino acids long, with a short helix. In the closed channel, the N-plug of each monomer was found to plug the bottom half of the neighboring pore, causing the channel to close. The most prominent loop L3 (labeled green) contained three short helices, which folded inside the central part of the protein pore. This loop was the most important loop in VhChiP, in that it contained several amino acid residues that bound to the sugar substrate (chitohexaose) (
      • Aunkham A.
      • Suginta W.
      Probing the physiological roles of the extracellular loops of chitoporin from Vibrio campbellii.
      ). Figure 2B (left panel) shows the side view of the monomeric SmChiP barrel, lacking the N-plug. Although loop L3 of SmChiP also protruded into the protein pore, it was not helical, leaving more space inside the channel interior (Fig. 2B, right panel). Superimposition of SmChiP onto one monomer of VhChiP yielded an R.M.S.D. of 3.75 Å for 940 atoms, reflecting large differences in the amino acid arrangement inside the two channels. We found a few residues that are conserved but were unable to identify the amino acid residues inside the SmChiP pore that could form the substrate binding sites, due to the high diversity in amino acid sequences between SmChiP and VhChiP.
      Figure thumbnail gr2
      Figure 2The overall structures of VhChiP and SmChiP. A, VhChiP. B, SmChiP. The structure of SmChiP was predicted by AlphaFold2 (https://alphafold.ebi.ac.uk/) as described in text, while the structure of VhChiP was retrieved from the PDB database (PDB id: 5MDQ). SmChiP, Serratia marcescens chitoporin.

      Determination of the molecular weight of native SmChiP

      The native state of the OM-expressed SmChiP was determined by size exclusion chromatography. Figure 3 shows the elution profiles of the protein standards together with SmChiP, using a HiPrep 26/60 prepacked Sephacryl S-300 column. SmChiP was eluted at a position between ovalbumin (43 kDa) and bovine serum albumin (66 kDa) (Fig. 3A). The eluted fractions obtained from the A280 peak were pooled and analyzed by SDS-PAGE. Figure 3A (inset) shows a Coomassie-stained protein band, which migrated at about 45 kDa, consistent with the apparent molecular mass of 50 kDa of SmChiP determined from the distribution coefficient (Kav) (Fig. 3B) and confirming that SmChiP is a monomer.
      Figure thumbnail gr3
      Figure 3Molecular weight determination by size-exclusion chromatography. A, standard proteins and SmChiP were resolved on a HiPrep 16/60 Sephacryl S-200 HR prepacked column under the conditions described in the text. B, the molecular weight of SmChiP was estimated from the plot of the distribution coefficient (Kav) versus log10 MW of four well-resolved standard proteins: ribonuclease, carbonic anhydrase, ovalbumin, and BSA. Control peaks were DNP-lysine, which was applied to determine the total volume of the GF column. The inset shows migration of SmChiP on SDS-PAGE, in comparison with that of the standard proteins. The protein bands were stained with Coomassie blue G-250 and photographed in monochrome. GF, gel filtration; SmChiP, Serratia marcescens chitoporin.

      Single-channel electrophysiology

      Single-channel measurements were carried out with lipid bilayer measurements (BLM) SmChiP reconstituted in 1,2-diphytanoyl-sn-glycero-3-phosphocholine membranes, and ion flow through the membrane-embedded SmChiP was recorded at different applied potentials. Figure 4, A and B show representative ion current traces acquired over 2000 ms at +100 mV and −100 mV, respectively. The channel was also found to be constantly open within a wide range of applied transmembrane potentials from ±25 mV to ±150 mV (data not shown).
      Figure thumbnail gr4
      Figure 4Pore-forming properties of SmChiP in artificial lipid membranes. Lipid bilayer measurements (BLM) were formed by a lowering-and-raising technique, using 5 mg ml−1 DPhPC bathed on either side with 1 M KCl in 20 mM Hepes, pH 7.4. The protein was always added to the cis side of the chamber. A, current trace at +100 mV, (B) current trace at −100 mV, and (C) multiple channel insertion. D, histogram analysis of channel conductance observed from 60 independent channel insertions. The distribution of the conductance profile was fitted to a Gaussian curve by Clampfit v.10.4. DPhPC, 1,2-diphytanoyl-sn-glycero-3-phosphatidylcholine; SmChiP, Serratia marcescens chitoporin.
      Ion traces acquired from a single SmChiP channel exhibited single-step openings with no subconductance or gating, at both negative and positive potentials. A current magnitude of approximately 50 pA was consistently observed under the applied potentials of ±100 mV in 1 M KCl electrolyte, giving the average single channel conductance of 0.5 nS (G = 0.5 nS). This conductance value of SmChiP is about one-third of that of trimeric VhChiP (G = 1.8 ± 0.3 nS) (
      • Suginta W.
      • Chumjan W.
      • Mahendran K.R.
      • Schulte A.
      • Winterhalter M.
      Chitoporin from Vibrio harveyi, a channel with exceptional sugar specificity.
      ), consistent with the monomeric structure of the SmChiP channel. Multichannel reconstitution experiments (Fig. 4C) further confirmed the mean channel conductance of 0.54 ± 0.2 nS from 60 independent channel insertions, obtained from Gaussian distribution fitting of the histogram (Fig. 4D). The channel conductance obtained from the I-V plot (Fig. 4E, for nine independent insertions) was 0.54 ± 0.01 nS, which was essentially identical with the value obtained from the multichannel insertions shown in Figure 4C.

      Sugar selectivity of SmChiP at single-channel level

      In this series of experiments, we examined the substrate specificity of SmChiP by exposing the channel to different chitooligosaccharides and examining their transient blocking behavior. In the absence of sugar, SmChiP was fully open, allowing a steady ionic current over the entire recording time of 2 min at both applied negative and positive potentials. Figure 5A shows a representative ion trace of an empty channel, which had ionic conductance of −50 pA at −100 mV (in Fig. 5 a trace of length 500-ms is shown). The addition of 80 μM D-GlcNAc (Fig. 5B) on either side of the chamber did not disturb the ion flow, while the addition of chitobiose at the same concentration (Fig. 5C) caused slight flickering of the current trace. Disturbance of the ion flow was seen when the channel was exposed to chitotriose. Figure 5D shows that the sugar molecules occluded the channel, causing short-lived blocking events throughout the recording time. The addition of chitotetraose (Fig. 5E) also yielded frequent, full blocking events. A similar blocking pattern was observed with chitopentaose, with more blocking events than were detected with chitotriose and chitotetraose (Fig. 5F). In the case of chitohexaose, although a lower number of blocking events was seen, they were full and long-lived (Fig. 5G). In contrast, addition of a nearly three-fold higher concentration (200 μM) of maltohexaose (Fig. 5H) or cellohexaose (Fig. 5I) did not interfere with ion current traces from SmChiP. The average residence time (τc) for channel blocking by chitotriose cannot be evaluated with confidence because of the limit of time resolution of our BLM instrument (<100 μS). However, the residence times for chitotetraose, chitopentaose, and chitohexaose were estimated to be 0.3, 1.3, and 6.0 ms, respectively.
      Figure thumbnail gr5
      Figure 5Channel specificity. Ion current fluctuations were monitored for 120 s at applied potentials of ±100 mV when sugar was added on either the cis or the trans side. Here, only current traces for 500 ms at −100 mV, with cis side addition, are shown. A, a fully open state of SmChiP before sugar addition. B, D-GlcNAc (N-acetylglucosamine), (C) chitobiose, (D) chitotriose, (E) chitotetraose, (F) chitopentaose, and (G) chitohexaose, added on the cis side of the chamber to a final concentration of 80 μM. H and I, are control recordings with maltohexaose and cellohexaose (each 200 μM), respectively. The histogram below each trace shows the distribution of the channel between open and closed (red arrows) state.
      Histogram analysis confirmed the blocking characteristics of individual sugar species (Fig. 5, AI, insets). D-GlcNAc, chitobiose, and chitotriose (Fig. 5, B–D, insets) did not alter the amplitude of the ionic current, which corresponded to the fully ‘open’ state of the channel (50 pA at +100 mV). On the other hand, transient blockings caused a discrete reduction of the ion current to zero when 80 μM of chitotetraose, chitopentaose, or chitohexaose was added. The fraction of the porin at zero current (labeled ‘closed’) increased as the chain length increased, indicating that long-chain chitooligosaccharides induced more complete blocking of the protein subunits (Fig. 5, E–G, insets).

      Bulk permeation of small monosaccharides and chitooligosaccharides through SmChiP

      Liposome swelling assays were carried out to examine the permeation of bulk sugars through the SmChiP-reconstituted proteoliposomes. Figure 6A shows the swelling rate with small sugar molecules of molecular weights 180 to 600 Da. The permeation rate of the tested sugar was estimated relative to that of the smallest sugar (D-arabinose, MW = 150), which was set to 100%. The isotonic concentration was determined to be 70 mM, using the impermeant sugar raffinose. All monosaccharides, including D-glucose, D-galactose, and D-mannose (all of MW = 180 kDa) and D-GlcNAc (MW = 222 Da), permeated the liposomes at relative rates of nearly 100%, while the disaccharides D-sucrose (MW = 342), D-maltose (MW = 360), and D-melezitose (MW = 522) were completely impermeant, reflecting a pore constriction limit of <300 Da. The permeation rates of different chitooligosaccharides were also tested. Figure 6B suggests that all chitooligosaccharides could permeate through the liposomes, even at a concentration as low as 2.5 mM. Chitotetraose, chitopentaose, and chitohexaose had slightly higher permeation rates than chitobiose and chitotriose. However, maltohexaose did not permeate through SmChiP at all.
      Figure thumbnail gr6
      Figure 6Proteoliposome swelling assays. D-raffinose was used to determine the isotonic concentration, i.e., the concentration of solute that produced no change in absorbance at 500 nm of the proteoliposome suspension over 60 s, and the swelling rate in L-arabinose was set to 100% to normalize swelling rates. The permeability of the channel was assumed to be proportional to the swelling rate. A, permeation of different types of small sugars (monosaccharides and disaccharides, 70 mM) through SmChiP reconstituted in liposomes. B, permeation of chitooligosaccharides (2 mM) through SmChiP. Values are means ± SD obtained from three independent sets of experiments. Circles on the bar graphs represent individual values. Statistical analysis was performed using One-way ANOVA, available in Prism. Significant differences are shown by asterisks (∗) and set at 0.001 < p < 0.05. SmChiP, Serratia marcescens chitoporin.

      Substrate binding affinity of SmChiP

      Isothermal microcalorimetric (ITC) titrations were carried out with SmChiP in solution with two preferred substrates (chitohexaose and chitopentaose) and were compared to the data with a nonchitin oligosaccharide (maltohexaose).
      Figure 7A shows the ITC thermograms obtained from titrating chitohexaose to SmChiP, and Figure 7B shows the theoretical fit of the normalized, integrated heat using a one-site binding model available in MicroCal PEAQ-ITC Analysis Software. Figure 7, C and D show the isothermal binding thermogram and the corresponding curve fit for chitopentaose. The equilibrium dissociation constant (Kd) was estimated to be 17 ± 0.6 μM for chitohexaose and 23 ± 2.0 μM for chitopentaose. Figure 7E is the isothermal titration profile of maltohexaose, showing no heat release (<0.1 μcal per injection) and yielding no integrated heat change (Fig. 7F), as a result of no binding.
      Figure thumbnail gr7
      Figure 7Binding studies of SmChiP with three oligosaccharides. Microcalorimetric titrations of SmChiP with oligosaccharides, showing ITC profiles corresponding to the binding of (A) chitohexaose, (C) chitopentaose, and (E) maltohexaose to SmChiP. B, integrated curve fitting for heat of binding of chitohexaose, (D) chitopentaose and (F) maltohexaose. ITC, isothermal microcalorimetric; SmChiP, Serratia marcescens chitoporin.

      Growth of S. marcescens on different carbon sources

      S. marcescens was grown on M9 minimal medium (MM) supplemented with three different carbon sources: D-glucose, a chitooligosaccharide mixture, and chitosan oligomers, and the growth was monitored at different time points from 0 to 96 h. Figure 8 shows the rapid growth of S. marcescens during the log phase of incubation (within 24 h), when D-glucose and chitin oligomers were used as the carbon source. The growth rate after entering the stationary phase declined less slowly with the cells grown on chitooligosaccharides than on D-glucose. S. marcescens could barely grow on MM supplemented with chitosan oligomers and was unable to grow on MM with no supplementary carbon source.
      Figure thumbnail gr8
      Figure 8Growth of S. marcescens on various carbon sources. S. marcescens were grown on M9 minimal medium supplemented with 0.5% (w/v) chitooligosaccharide mixture, D-glucose, or chitosan oligomers. The growth rate was monitored at different time points of incubation up to 96 h, at 26 °C.

      Discussion

      The chiP gene encoding SmChiP from S. marcescens 2170 was first identified as part of the ybfMN-ctp gene cluster. Later, this was referred to as the chiPQ-ctb cluster (
      • Toratani T.
      • Suzuki K.
      • Shimizu M.
      • Sugimoto H.
      • Watanabe T.
      Regulation of chitinase production by the 5'-untranslated region of the ybfM in Serratia marcescens 2170.
      ). The same study also generated the S. marcescens mutants lacking the chiP, chiX, and chiQ genes. Deletions of these target genes affected the growth of the bacterium. In particular, the △chiP mutant showed significantly lower ability to grow on medium supplemented with colloidal chitin and (GlcNAc)2, and no growth on medium containing (GlcNAc)3. Subsequent studies demonstrated that the chiX small RNA controlled the expression of chitin-degrading enzymes (chitinases, chitobiase, and N-acetylglucosaminidase), chitin-binding protein (CBP21) (
      • Suzuki K.
      • Suzuki M.
      • Taiyoji M.
      • Nikaidou N.
      • Watanabe T.
      Chitin binding protein (CBP21) in the culture supernatant of Serratia marcescens 2170.
      ), and a chitooligosaccharide-transporting porin (later named SmChiP) (
      • Suginta W.
      • Chumjan W.
      • Mahendran K.R.
      • Janning P.
      • Schulte A.
      • Winterhalter M.
      Molecular uptake of chitooligosaccharides through chitoporin from the marine bacterium Vibrio harveyi.
      ,
      • Toratani T.
      • Suzuki K.
      • Shimizu M.
      • Sugimoto H.
      • Watanabe T.
      Regulation of chitinase production by the 5'-untranslated region of the ybfM in Serratia marcescens 2170.
      ). Nonetheless, no detailed functional and structural characterization of SmChiP has been reported to date. In the present study, we heterologously expressed SmChiP in the E. coli system. The pore-forming properties of SmChiP in lipid membranes were elucidated at the single-molecule level. In contrast to the well-characterized trimeric VhChiP from V. campbellii (formerly V. harveyi) (
      • Suginta W.
      • Chumjan W.
      • Mahendran K.R.
      • Janning P.
      • Schulte A.
      • Winterhalter M.
      Molecular uptake of chitooligosaccharides through chitoporin from the marine bacterium Vibrio harveyi.
      ,
      • Aunkham A.
      • Suginta W.
      Probing the physiological roles of the extracellular loops of chitoporin from Vibrio campbellii.
      ,
      • Toratani T.
      • Suzuki K.
      • Shimizu M.
      • Sugimoto H.
      • Watanabe T.
      Regulation of chitinase production by the 5'-untranslated region of the ybfM in Serratia marcescens 2170.
      ) and VcChiP from Vibrio cholerae (
      • Soysa H.S.M.
      • Aunkham A.
      • Schulte A.
      • Suginta W.
      Single-channel properties, sugar specificity, and role of chitoporin in adaptive survival of Vibrio cholerae type strain O1.
      ), SmChiP was shown to be a member of the class of monomeric porins, like the closely related EcChiP (
      • Soysa H.S.M.
      • Schulte A.
      • Suginta W.
      Functional analysis of an unusual porin-like channel that imports chitin for alternative carbon metabolism in Escherichia coli.
      ). Although most porins form trimers, a few were shown to be monomeric, including Pseudomonas aeruginosa OCCDs and OCCKs (formerly referred as OprD porins) (
      • Biswas S.
      • Mohammad M.M.
      • Patel D.R.
      • Movileanu L.
      • van den Berg B.
      Structural insight into OprD substrate specificity.
      ,
      • Eren E.
      • Parkin J.
      • Adelanwa A.
      • Cheneke B.
      • Movileanu L.
      • Khalid S.
      • et al.
      Toward understanding the outer membrane uptake of small molecules by Pseudomonas aeruginosa.
      ) and E. coli OmpG (
      • Yildiz O.
      • Vinothkumar K.V.
      • Goswami P.
      • Kühlbrandt W.
      Structure of the monomeric outer-membrane porin OmpG in the open and closed conformation.
      ).
      Bulk permeability of different sugars through SmChiP, as determined by a liposome swelling assay, provided an estimated size exclusion limit of about 300 Da. In our study, only monosaccharides were able to pass through the protein pore by general diffusion. SmChiP exhibited clear substrate specificity, being completely impermeable to nonchitin oligosaccharides. However, BLM results showed that chitobiose and longer-chain chitooligosaccharides (chitotriose, -tetraose, -pentaose and -hexaose) could permeate through SmChiP, despite their MWs exceeding its size exclusion limit. We recently showed that the C2-acetamido functionality on the GlcNAc units of the chitooligosaccharide chain served as a molecular footprint for sugar-channel recognition (
      • Suginta W.
      • Sanram S.
      • Aunkham A.
      • Winterhalter M.
      • Schulte A.
      The C2 entity of chitosugars is crucial in molecular selectivity of the Vibrio campbellii chitoporin.
      ). SmChiP was shown to have no binding affinity for chitosan oligomers (deacetylated chitooligosaccharides), since these sugars lack the C2-acetamido functional groups. Our thermodynamic data obtained from ITC experiments confirmed that the binding affinity for the most highly preferred substrate, chitohexaose (Kd = 17 μM or K = 60,000 M−1) was 4.2-fold and 8.4-fold lower than for chitohexaose binding to EcChiP (K = 250,000 M−1) (
      • Soysa H.S.M.
      • Schulte A.
      • Suginta W.
      Functional analysis of an unusual porin-like channel that imports chitin for alternative carbon metabolism in Escherichia coli.
      ) and to VhChiP (K = 500,000 M−1), respectively. The ITC data confirmed that SmChiP had weaker substrate-binding affinity than EcChiP (
      • Soysa H.S.M.
      • Schulte A.
      • Suginta W.
      Functional analysis of an unusual porin-like channel that imports chitin for alternative carbon metabolism in Escherichia coli.
      ) and VhChiP (
      • Toratani T.
      • Suzuki K.
      • Shimizu M.
      • Sugimoto H.
      • Watanabe T.
      Regulation of chitinase production by the 5'-untranslated region of the ybfM in Serratia marcescens 2170.
      ).
      Single-channel recordings suggested that SmChiP could form a highly stable channel over a wide range of applied external potentials (±25 to ±150 mV), with neither gating nor stepwise closure. The average single-channel conductance of SmChiP (G = 0.5 ± 0.2 nS) was approximately one-third of the mean conductance reported for trimeric VhChiP (G = 1.8 ± 0.3 nS) (
      • Toratani T.
      • Suzuki K.
      • Shimizu M.
      • Sugimoto H.
      • Watanabe T.
      Regulation of chitinase production by the 5'-untranslated region of the ybfM in Serratia marcescens 2170.
      ), which confirmed the monomeric state of SmChiP. Cell growth assays showed that S. marcescens could grow on chitin oligosaccharides, owing to its ability to digest chitinous materials and transport the degradation products through SmChiP into the cell.
      Figure 9 is a schematic illustration of chitin processing by the OM and inner membrane (IM) of S. marcescens, suggested by Watanabe et al. (
      • Uchiyama T.
      • Kaneko R.
      • Yamaguchi J.
      • Inoue A.
      • Yanagida T.
      • Nikaidou N.
      • et al.
      Uptake of N,N′-diacetylchitobiose [(GlcNAc)2] via the phosphotransferase system is essential for chitinase production by Serratia marcescens 2170.
      ,
      • Suzuki K.
      • Shimizu M.
      • Sasaki N.
      • Ogawa C.
      • Minami H.
      • Sugimoto H.
      • et al.
      Regulation of the chitin degradation and utilization system by the ChiX small RNA in Serratia marcescens 2170.
      ). Chitin is initially digested by chitinases, generating chitooligosaccharides of various lengths as the initial products, and finally further generating (GlcNAc)2 as the major product (dark arrow), along with the monosaccharide (D-GlcNAc) as a minor product (dashed arrow). The chitin degradation products enter the periplasm by two possible routes: (GlcNAc)2 and chitooligosaccharides through SmChiP and GlcNAc through a general diffusion porin. (GlcNAc)2 in the periplasm was suggested to be transported further through the IM by a specific (GlcNAc)2-specific enzyme IIC, encoded by the chb operon (
      • Uchiyama T.
      • Kaneko R.
      • Yamaguchi J.
      • Inoue A.
      • Yanagida T.
      • Nikaidou N.
      • et al.
      Uptake of N,N′-diacetylchitobiose [(GlcNAc)2] via the phosphotransferase system is essential for chitinase production by Serratia marcescens 2170.
      ). On the other hand, GlcNAc is further transported through the IM by a GlcNAc-specific phosphoenolpyruvate:carbohydrate phosphotransferase system transporter. (GlcNAc)2 and GlcNAc entering the cytoplasm via phosphoenolpyruvate:carbohydrate phosphotransferase system are phosphorylated during transport to (GlcNAc)2-P and GlcNAc-6P, which are further metabolized in the cytoplasm (
      • Toratani T.
      • Shoji T.
      • Ikehara T.
      • Suzuki K.
      • Watanabe T.
      The importance of chitobiase and N-acetylglucosamine (GlcNAc) uptake in N,N'-diacetylchitobiose [(GlcNAc)2] utilization by Serratia marcescens 2170.
      ).
      Figure thumbnail gr9
      Figure 9Chitin utilization by S. marcescens. GlcNAc and (GlcNAc)2–6 are chitooligosaccharides generated by the cleavage of chitin by chitinase. GlcNAcase is a periplasmic β-N-acetylglucosaminidase and GlcN is glucosamine. Gray arrows indicate the main enzymic route, while dashed arrows indicate minor enzymic routes. Red arrows indicate translocation steps. IM, inner membrane; OM, outer membrane; PTS, phosphoenolpyruvate:carbohydrate phosphotransferase system.
      In conclusion, this study elucidates the physiological function of a chitooligosaccharide-specific porin from the Serratia system and provides an understanding of how the Serratia bacteria can utilize chitin as their carbon source. Given that SmChiP is the molecular gateway for nutrient uptake, it may serve as an excellent protein target for the strategic design of effective antimicrobial agents with a novel mode of action, such as the chemoenzymatic synthesis of chitooligosaccharide-based compounds that can inhibit Serratia infections.

      Experimental procedures

      Bacterial strains and vectors

      The pET23a(+) expression vector, carrying the S. marcescens ChiP gene, was obtained from GenScript USA Inc. E. coli strain DH5α, used for plasmid preparations, was obtained from Invitrogen (Gibthai Company, Ltd). E. coli BL21 (DE3) Omp8 Rosetta strain, lacking the major endogenous Omps (OmpF, OmpC, OmpA, and LamB) was a gift from Professor Dr Roland Benz, Jacobs University Bremen, Germany. Chitooligosaccharides were purchased from Dextra Laboratories and Megazyme.

      Structure-based sequence alignment and AlphaFold2 structural prediction

      Structure-based sequence alignment was performed for representative ChiPs from S. marcescens (SmChiP; protein id: A0A0PBS3), E. coli (EcChiP, protein id: P75733), V. cholerae (VcChiP, protein id: Q9KTD0), and V. harveyi, now classified as V. campbellii (VhChiP, protein id: L0RYU0). The corresponding sequences were retrieved from the UniProtKB database (https://www.uniprot.org/uniprot/), aligned by Muscle alignment (
      • Katoh K.
      • Rozewicki J.
      • Yamada K.D.
      MAFFT online service: multiple sequence alignment, interactive sequence choice and visualization.
      ) and displayed in Jalview v. 2.11.2.1. The 2D structural elements of monomeric SmChiP were constructed from AlphaFold2 (https://alphafold.ebi.ac.uk/), while the 2D structural elements of trimeric VhChiP were constructed from the crystal structure of the native form (PDB id: 5MDO).

      Recombinant expression, cell wall extraction, and protein purification

      Cloning of the chiP gene encoding SmChiP and purification of the recombinant protein were described in our previous report (
      • Amornloetwattana R.
      • Robinson R.C.
      • Soysa H.S.M.
      • van den Berg B.
      • Suginta W.
      Chitoporin from Serratia marcescens: recombinant expression, purification and crystallization.
      ). Briefly, the overnight culture of transformed cells was transferred to Luria-Bertani (LB) broth containing 100 μg ml−1 ampicillin and 25 μg ml−1 kanamycin and grown at 25 + 1 °C until A600 ∼0.6 to 0.8. SmChiP expression was then induced with 0.5 mM of isopropyl thio-β-D-galactoside for 6 h at 37 °C. For protein purification, the cell pellet, collected after centrifugation and resuspended in lysis buffer (20 mM Tris-HCl pH 8.0, 2.5 mM MgCl2, 0.1 mM CaCl2), containing 10 μg ml−1 RNase A and 10 μg ml−1 DNase I, was subjected to high-speed ultrasonic homogenization (EmulsiFlex-C3). Cell walls were extracted by incubating the crude extract in 2% (w/v) SDS solution at 50 °C for 60 min. After centrifugation at 100,000g at 4 °C for 1 h, the cell wall faction (pellet) was extracted twice with 2.5% (v/v) octyl-POE (n-octylpolyoxyethylene; ALEXIS Biochemicals). After centrifugation at the same speed, the supernatant containing OM-expressed SmChiP was dialyzed thoroughly against 20 mM potassium phosphate buffer, pH 7.4, containing 0.05% (v/v) lauryldimethylamine oxide (Sigma-Aldrich Pte Ltd), and purified by ion-exchange chromatography on a Hitrap Q HP prepacked column (1 cm Ø × 5 cm L), followed by gel filtration chromatography on a HiPrep 16/60 Sephacryl S-200 High Resolution column (GE Healthcare Life Sciences). The purity of the SmChiP fractions obtained after the gel filtration step was verified by SDS-PAGE, and the protein concentration of the purified SmChiP was estimated using the Novagen BCA protein assay kit (EMD Chemicals Inc)

      Molecular weight determination of native SmChiP

      The molecular weight of SmChiP in its native state was determined by size-exclusion chromatography. Standard proteins of known molecular weight were resolved on a HiPrep 16/60 Sephacryl S-200 HR prepacked column (GE Healthcare Life Sciences) under the conditions described above. Dextran-2000 was used to obtain the void volume (V0), while DNP-lysine was used to calculate the volume of the stationary phase (Vi). With the elution volume of each protein sample denoted Ve, the elution of the protein sample is described by the distribution coefficient (Kav) defined in Equation 1 (
      • Tayyab S.
      • Qamar S.
      • Islam M.
      Size exclusion chromatography and size exclusion HPLC of proteins.
      ):
      Kav=VeVoVi
      (1)


      A calibration curve was created by plotting Kav versus logarithmic values of the corresponding molecular weights of the standard proteins and was used to estimate the molecular weight of SmChiP. The standard proteins used in this experiment were ferritin (440 kDa), catalase (250 kDa), aldolase (158 kDa), bovine serum albumin (66 kDa), ovalbumin (43 kDa), carbonic anhydrase (29 kDa), and ribonuclease A (13.7 kDa).

      Single-channel electrophysiology

      Montal-Mueller type solvent-free bilayer (
      • Montal M.
      • Mueller P.
      Formation of bimolecular membranes from lipid monolayers and a study of their electrical properties.
      ) formation was performed using 5 mg ml−1 1,2-diphytanoyl-sn-glycero-3-phosphatidylcholine (Avanti Polar Lipids) in n-pentane. First, a 25-mm-thick Teflon film with an aperture of 50 to 100 mm was sandwiched between the two chambers of a cuvette, and the aperture was prepainted with a few microliters of 1% (v/v) hexadecane in hexane. The chambers were filled with 1 M KCl in 20 mM Hepes, pH 7.4, and BLM experiments were carried out at 25 °C. A planar bilayer was formed across the aperture by lowering and raising the liquid level (
      • Suginta W.
      • Chumjan W.
      • Mahendran K.R.
      • Janning P.
      • Schulte A.
      • Winterhalter M.
      Molecular uptake of chitooligosaccharides through chitoporin from the marine bacterium Vibrio harveyi.
      ). Ionic currents were detected using Ag/AgCl electrodes, with the reference electrode connected to the cis side of the membrane (ground) and the working electrode connected to the head-stage of an Axopatch 200B amplifier (Axon Instruments). Single-channel measurements were performed in the voltage clamp mode and digitized using the Axon Digidata 1550 digitizer, and the data acquisition was performed using Clampex software (Axon Instruments). Using a low-pass Bessel filter with a sampling frequency of 50 kHz, the traces obtained were filtered at 10 kHz. Single channel analyses were performed using Clampfit software (all from Molecular Devices). Single protein channels were reconstituted in lipid, and fully open channels of SmChiP were titrated with discrete concentrations of chitooligosaccharides at the cis or trans side of the chamber. Fluctuations of ion flow produced by sugar diffusion through the inserted channel were usually recorded for 2 min at different transmembrane potentials of ±25 to ±100 mV. Multiple channel insertion experiments were operated with a patch-clamp amplifier connected to a two-electrode bilayer head-stage (PC-ONE plus PC-ONE-50; Dagan Corp) together with an A/D converter (LIH 1600, HEKA Elektronik) that was operated using PULSE program (HEKA Elektronik).

      Liposome swelling experiments

      The SmChiP-reconstituted proteoliposomes were prepared as described elsewhere (
      • Soysa H.S.M.
      • Schulte A.
      • Suginta W.
      Functional analysis of an unusual porin-like channel that imports chitin for alternative carbon metabolism in Escherichia coli.
      ,
      • Luckey M.
      • Nikaido H.
      Specificity of diffusion channels produced by λ phage receptor protein of Escherichia coli.
      ). Soybean l-α-phosphatidylcholine [20 mg ml−1, (Sigma) freshly prepared in chloroform] was used to form multilamellar liposomes. For the preparation of proteoliposomes, 200 ng of SmChiP was reconstituted into 200 μl of the liposome suspension by sonication, and then 17% (w/v) dextran (40 kDa) was entrapped in the proteoliposomes. D-Raffinose solutions were prepared in 20 mM potassium phosphate buffer, pH 7.4, to obtain concentrations of 40, 50, 60, and 70 mM for determination of the isotonic solute concentration. This value was then used for the adjustment of the isotonic concentration for other solutes. To carry out a liposome-swelling assay, 25 μl of the proteoliposome suspension was added to 600 μl of sugar solution, and changes in absorbance at 500 nm were monitored immediately. The apparent absorbance change over the first 60 s was used to estimate the swelling rate (s−1) following the equation φ = (1/Ai)dA/dt in which φ is the swelling rate, Ai is the initial absorbance, and dA/dt is the rate of absorbance change during the first 60 s. The swelling rate for each sugar was normalized by setting the rate of L-arabinose (150 Da) to 100%. The values presented are averages from three independent determinations. Protein-free liposomes and proteoliposomes without sugars were used as negative controls. The sugars tested were D-glucose (180 Da), D-mannose (180 Da), D-galactose (180 Da), N-acetylglucosamine (GlcNAc) (221 Da), D-sucrose (342 Da), D-melezitose (522 Da), GlcNAc2 (424 Da), GlcNAc3 (628 Da), GlcNAc4 (830 Da), GlcNAc5 (1034 Da), GlcNAc6 (1237 Da), and maltodextrins.

      Chitooligosaccharide-induced growth of S. marcescens

      S. marcescens was streaked onto LB agar plates, then incubated at 26 °C overnight. A single colony was picked and inoculated into LB medium for gentle shaking at 26 °C overnight. The bacterial cells were grown in M9 minimal medium (26 mM Na2HPO4, 22 mM KH2PO4, 19 mM NH4Cl, 2 mM MgSO4, 0.1 mM CaCl2, and 8.6 mM NaCl), supplemented with various carbon sources (D-glucose, chitosan oligomers or chitooligosaccharides), all at 0.5% (w/v). Cells (150 μl) were collected to measure A600 in a microtiter plate during each time course (0–96 h). Cell growth was monitored with a microplate reader model MULTISKAN Sky (Thermo Fisher Scientific). The number of cells (cfu.ml−1) was calculated from A600 (0.1 A600 = 1 × 108 cfu.ml−1) (
      • Szewzyk Z.
      • Szewzyk R.
      • Stenström T.A.
      Growth and survival of Serratia marcescens under aerobic and anaerobic conditions in the presence of materials from blood bags.
      ).

      Data availability

      The BLM and ITC datasets for this study can be found in the following URL link: https://drive.google.com/drive/folders/1qh5qDiOhLt0-c18cgT41XgZ42AY8jyA4.

      Conflict of interest

      The authors declare that they have no conflicts of interest with the contents of this article.

      Acknowledgments

      We would like to acknowledge the help of Dr David Apps, University of Edinburgh, Scotland for critical readings of the manuscript. We would also like to thank Professor Mathias Winterhalter, Jacobs University Bremen, Germany, for providing the BLM setup for single channel measurements.

      Author contributions

      H. S. M. S., S. K., and R. A. methodology; H. S. M. S., S. K., and R. A. software; H. S. M. S., S. K., and R. A. data curation; H. S. M. S., S. K., and R. A. writing-original draft; H. S. M. S., S. K., and R. A. visualization; H. S. M. S., S. K., and R. A. investigation; H. S. M. S. formal analysis; T. W. and W. S. conceptualization; T. W. and W. S. writing-review and editing; W. S. supervision; W. S. validation.

      Funding and additional information

      This research was funded by Thailand Science Research and Innovation (TSRI) and Vidyasirimedhi Institute of Science and Technology (VISTEC) ( BRG610008 ). H. S. M. S. received PhD funding from Suranaree University of Technology (SUT) through an OGRG grant.

      References

        • Mahlen S.D.
        Serratia infections: from military experiments to current practice.
        Clin. Microbiol. Rev. 2011; 24: 755-791
        • Yoon H.J.
        • Choi J.Y.
        • Park Y.S.
        • Kim C.O.
        • Kim J.M.
        • Yong D.E.
        • et al.
        Outbreaks of Serratia marcescens bacteriuria in a neurosurgical intensive care unit of a tertiary care teaching hospital: a clinical, epidemiologic, and laboratory perspective.
        Am. J. Infect. Control. 2005; 33: 595-601
        • Voelz A.
        • Müller A.
        • Gillen J.
        • Le C.
        • Dresbach T.
        • Engelhart S.
        • et al.
        Outbreaks of Serratia marcescens in neonatal and pediatric intensive care units: clinical aspects, risk factors and management.
        Int. J. Hyg. Environ. Health. 2009; 213: 79-87
        • Jones R.N.
        Microbial etiologies of hospital-acquired bacterial pneumonia and ventilator-associated bacterial pneumonia.
        Clin. Infect. Dis. 2010; 51: S81-S87
        • van der Vorm E.R.
        • Woldring-Zwaan C.
        Source, carriers, and management of a Serratia marcescens outbreak on a pulmonary unit.
        J. Hosp. Infect. 2002; 52: 263-267
        • Aronson L.C.
        • Alderman S.I.
        The occurrence and bacteriological characteristics of S. marcescens from a case of meningitis.
        J. Bacteriol. 1943; 43: 261-267
        • Rabinowitz K.
        • Schiffrin R.
        A ward contamination by Serratia marcescens.
        Acta Med. Orient. 1952; 11: 181-184
        • Kawecki D.
        • Kwiatkowski A.
        • Sawicka-Grzelak A.
        • Durlik M.
        • Paczek L.
        • Chmura A.
        • et al.
        Urinary tract infections in the early posttransplant period after kidney transplantation: etiologic agents and their susceptibility.
        Transplant. Proc. 2011; 43: 2991-2993
        • Nikaido H.
        Outer membrane barrier as a mechanism of antimicrobial resistance.
        Antimicrob. Agents Chemother. 1989; 33: 1831-1836
        • Stock I.
        • Grueger T.
        • Wiedemann B.
        Natural antibiotic susceptibility of strains of Serratia marcescens and the S. liquefaciens complex: S. liquefaciens sensu stricto, S. proteamaculans and S. grimesii.
        Int. J. Antimicrob. Agents. 2003; 22: 35-47
        • Arakawa Y.
        Systematic research to overcome newly emerged multidrug-resistant bacteria.
        Microbiol. Immunol. 2020; 64: 231-251
        • Cooksey R.C.
        • Thorne G.M.
        • Farrar Jr., W.E.
        R factor-mediated antibiotic resistance in Serratia marcescens.
        Antimicrob. Agents Chemother. 1976; 10: 123-127
        • Shirshikova T.V.
        • Sierra-Bakhshi C.G.
        • Kamaletdinova L.K.
        • Matrosova L.E.
        • Khabipova N.N.
        • Evtugyn V.G.
        • et al.
        The ABC-type efflux pump MacAB is involved in protection of Serratia marcescens against aminoglycoside antibiotics, polymyxins, and oxidative stress.
        mSphere. 2021; 6e00033-21
        • Hall-Stoodley L.
        • Costerton J.W.
        • Stoodley P.
        Bacterial biofilms: from the natural environment to infectious diseases.
        Nat. Rev. Microbiol. 2004; 2: 95-108
        • Satpathy S.
        • Sen S.K.
        • Pattanaik S.
        • Raut S.
        Review on bacterial biofilm: an universal cause of contamination.
        Biocatal. Agric. Biotechnol. 2016; 7: 56-66
        • Srinivasan R.
        • Devi K.R.
        • Kannappan A.
        • Pandian S.K.
        • Ravi A.V.
        Piper betle and its bioactive metabolite phytol mitigates quorum sensing mediated virulence factors and biofilm of nosocomial pathogen Serratia marcescens in vitro.
        J. Ethnopharmacol. 2016; 193: 592-603
        • Ray C.
        • Shenoy A.T.
        • Orihuela C.J.
        • González-Juarbe N.
        Killing of Serratia marcescens biofilms with chloramphenicol.
        Ann. Clin. Microbiol. Antimicrob. 2017; 16: 19
        • Jacoby G.A.
        AmpC beta-lactamases.
        Clin. Microbiol. Rev. 2009; 22: 161-182
        • Bes T.
        • Nagano D.
        • Martins R.
        • Marchi A.P.
        • Perdigão-Neto L.
        • Higashino H.
        • et al.
        Bloodstream infections caused by Klebsiella pneumoniae and Serratia marcescens isolates co-harboring NDM-1 and KPC-2.
        Ann. Clin. Microbiol. Antimicrob. 2021; 20: 57
        • Harris P.N.
        • Ferguson J.K.
        Antibiotic therapy for inducible AmpC beta-lactamase-producing Gram-negative bacilli: what are the alternatives to carbapenems, quinolones and aminoglycosides?.
        Int. J. Antimicrob. Agents. 2012; 40: 297-305
        • Vaaje-Kolstad G.
        • Horn S.J.
        • Sørlie M.
        • Eijsink V.G.H.
        The chitinolytic machinery of Serratia marcescens--a model system for enzymatic degradation of recalcitrant polysaccharides.
        FEBS J. 2013; 280: 3028-3049
        • Uchiyama T.
        • Kaneko R.
        • Yamaguchi J.
        • Inoue A.
        • Yanagida T.
        • Nikaidou N.
        • et al.
        Uptake of N,N′-diacetylchitobiose [(GlcNAc)2] via the phosphotransferase system is essential for chitinase production by Serratia marcescens 2170.
        J. Bacteriol. 2003; 185: 1776-1782
        • Suzuki K.
        • Shimizu M.
        • Sasaki N.
        • Ogawa C.
        • Minami H.
        • Sugimoto H.
        • et al.
        Regulation of the chitin degradation and utilization system by the ChiX small RNA in Serratia marcescens 2170.
        Biosci. Biotechnol. Biochem. 2016; 80: 376-385
        • Monreal J.
        • Reese E.T.
        The chitinase of Serratia marcescens.
        Can. J. Microbiol. 1969; 15: 689-696
        • Watanabe T.
        • Kimura K.
        • Sumiya T.
        • Nikaidou N.
        • Suzuki K.
        • Suzuki M.
        • et al.
        Genetic analysis of the chitinase system of Serratia marcescens 2170.
        J. Bacteriol. 1997; 179: 7111-7117
        • Suzuki K.
        • Sugawara N.
        • Suzuki M.
        • Uchiyama T.
        • Katouno F.
        • Nikaidou N.
        • et al.
        Chitinases A, B, and C1 of Serratia marcescens 2170 produced by recombinant Escherichia coli: enzymatic properties and synergism on chitin degradation.
        Biosci. Biotechnol. Biochem. 2002; 66: 1075-1083
        • Takanao S.
        • Honma S.
        • Miura T.
        • Ogawa C.
        • Sugimoto H.
        • Suzuki K.
        • et al.
        Construction and basic characterization of deletion mutants of the genes involved in chitin utilization by Serratia marcescens 2170.
        Biosci. Biotechnol. Biochem. 2014; 78: 524-532
        • Synstad B.
        • Vaaje-Kolstad G.
        • Cederkvist F.H.
        • Saua S.F.
        • Horn S.J.
        • Eijsink V.G.H.
        • et al.
        Expression and characterization of endochitinase C from Serratia marcescens BJL200 and its purification by a one-step general chitinase purification method.
        Biosci. Biotechnol. Biochem. 2008; 72: 715-723
        • Sánchez L.
        • Ruiz N.
        • Leranoz S.
        • Viñas M.
        • Puig M.
        The role of outer membrane in Serratia marcescens intrinsic resistance to antibiotics.
        Microbiologia. 1997; 13: 315-320
        • Begic S.
        • Worobec E.A.
        Regulation of Serratia marcescens ompF and ompC porin genes in response to osmotic stress, salicylate, temperature and pH.
        Microbiology (Reading). 2006; 152: 485-491
        • Amornloetwattana R.
        • Robinson R.C.
        • Soysa H.S.M.
        • van den Berg B.
        • Suginta W.
        Chitoporin from Serratia marcescens: recombinant expression, purification and crystallization.
        Acta Crystallogr. F Struct. Biol. Commun. 2020; 76: 536-543
        • Keyhani N.O.
        • Li X.B.
        • Roseman S.
        Chitin catabolism in the marine bacterium Vibrio furnissii. Identification and molecular cloning of a chitoporin.
        J. Biol. Chem. 2000; 275: 33068-33076
        • Suginta W.
        • Chumjan W.
        • Mahendran K.R.
        • Janning P.
        • Schulte A.
        • Winterhalter M.
        Molecular uptake of chitooligosaccharides through chitoporin from the marine bacterium Vibrio harveyi.
        PLoS One. 2013; 8e55126
        • Soysa H.S.M.
        • Aunkham A.
        • Schulte A.
        • Suginta W.
        Single-channel properties, sugar specificity, and role of chitoporin in adaptive survival of Vibrio cholerae type strain O1.
        J. Biol. Chem. 2020; 295: 9421-9432
        • Soysa H.S.M.
        • Schulte A.
        • Suginta W.
        Functional analysis of an unusual porin-like channel that imports chitin for alternative carbon metabolism in Escherichia coli.
        J. Biol. Chem. 2017; 292: 19328-19337
        • Soysa H.S.M.
        • Suginta W.
        • Moonsap W.
        • Smith M.F.
        Chitosugar translocation by an unexpressed monomeric protein channel.
        Phys. Rev. E. 2018; 97: 052417
        • Aunkham A.
        • Zahn M.
        • Kesireddy A.
        • Pothula K.R.
        • Schulte A.
        • Baslé A.
        • et al.
        Structural basis for chitin acquisition by marine Vibrio species.
        Nat. Commun. 2018; 9: 220
        • Aunkham A.
        • Suginta W.
        Probing the physiological roles of the extracellular loops of chitoporin from Vibrio campbellii.
        Biophys. J. 2021; 120: 2124-2137
        • Suginta W.
        • Chumjan W.
        • Mahendran K.R.
        • Schulte A.
        • Winterhalter M.
        Chitoporin from Vibrio harveyi, a channel with exceptional sugar specificity.
        J. Biol. Chem. 2013; 288: 11038-11046
        • Toratani T.
        • Suzuki K.
        • Shimizu M.
        • Sugimoto H.
        • Watanabe T.
        Regulation of chitinase production by the 5'-untranslated region of the ybfM in Serratia marcescens 2170.
        Biosci. Biotechnol. Biochem. 2012; 76: 1920-1924
        • Suzuki K.
        • Suzuki M.
        • Taiyoji M.
        • Nikaidou N.
        • Watanabe T.
        Chitin binding protein (CBP21) in the culture supernatant of Serratia marcescens 2170.
        Biosci. Biotechnol. Biochem. 1998; 62: 128-135
        • Biswas S.
        • Mohammad M.M.
        • Patel D.R.
        • Movileanu L.
        • van den Berg B.
        Structural insight into OprD substrate specificity.
        Nat. Struct. Mol. Biol. 2007; 14: 1108-1109
        • Eren E.
        • Parkin J.
        • Adelanwa A.
        • Cheneke B.
        • Movileanu L.
        • Khalid S.
        • et al.
        Toward understanding the outer membrane uptake of small molecules by Pseudomonas aeruginosa.
        J. Biol. Chem. 2013; 288: 12042-12053
        • Yildiz O.
        • Vinothkumar K.V.
        • Goswami P.
        • Kühlbrandt W.
        Structure of the monomeric outer-membrane porin OmpG in the open and closed conformation.
        EMBO J. 2006; 25: 3702-3713
        • Suginta W.
        • Sanram S.
        • Aunkham A.
        • Winterhalter M.
        • Schulte A.
        The C2 entity of chitosugars is crucial in molecular selectivity of the Vibrio campbellii chitoporin.
        J. Biol. Chem. 2021; 297: 101350
        • Toratani T.
        • Shoji T.
        • Ikehara T.
        • Suzuki K.
        • Watanabe T.
        The importance of chitobiase and N-acetylglucosamine (GlcNAc) uptake in N,N'-diacetylchitobiose [(GlcNAc)2] utilization by Serratia marcescens 2170.
        Microbiology (Reading). 2008; 154: 1326-1332
        • Katoh K.
        • Rozewicki J.
        • Yamada K.D.
        MAFFT online service: multiple sequence alignment, interactive sequence choice and visualization.
        Brief. Bioinform. 2019; 20: 1160-1166
        • Tayyab S.
        • Qamar S.
        • Islam M.
        Size exclusion chromatography and size exclusion HPLC of proteins.
        Biochem. Educ. 1991; 19: 149-152
        • Montal M.
        • Mueller P.
        Formation of bimolecular membranes from lipid monolayers and a study of their electrical properties.
        Proc. Natl. Acad. Sci. U. S. A. 1972; 69: 3561-3566
        • Luckey M.
        • Nikaido H.
        Specificity of diffusion channels produced by λ phage receptor protein of Escherichia coli.
        Proc. Natl. Acad. Sci. U. S. A. 1980; 77: 167-171
        • Szewzyk Z.
        • Szewzyk R.
        • Stenström T.A.
        Growth and survival of Serratia marcescens under aerobic and anaerobic conditions in the presence of materials from blood bags.
        J. Clin. Microbiol. 1993; 31: 1826-1830